{"id":29895,"date":"2019-10-31T21:32:30","date_gmt":"2019-10-31T18:32:30","guid":{"rendered":"https:\/\/prohoster.info\/blog\/nejroset-ot-nvidia-prevrashhaet-prostejshie-nabroski-v-krasivye-pejzazhi\/"},"modified":"2019-10-31T21:32:30","modified_gmt":"2019-10-31T18:32:30","slug":"nejroset-ot-nvidia-prevrashhaet-prostejshie-nabroski-v-krasivye-pejzazhi","status":"publish","type":"post","link":"https:\/\/prohoster.info\/cs\/blog\/nejroset-ot-nvidia-prevrashhaet-prostejshie-nabroski-v-krasivye-pejzazhi","title":{"rendered":"Neuros\u00ed\u0165 od Nvidie prom\u011b\u0148uje jednoduch\u00e9 n\u00e1\u010drty na kr\u00e1sn\u00e1 krajinn\u00e1 um\u011bn\u00ed.","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

\"Neuros\u00ed\u0165
\nVodop\u00e1d ku\u0159\u00e1ka a vodop\u00e1d zdrav\u00e9ho \u010dlov\u011bka<\/p>\n

V\u0161ichni v\u00edme, jak nakreslit sovu. Nejprve je t\u0159eba nakreslit ov\u00e1l, pak je\u0161t\u011b jednu kru\u017enici, a nakonec - vznikne n\u00e1dhern\u00e1 sova. Samoz\u0159ejm\u011b, to je vtip, a to velmi star\u00fd, ale in\u017een\u00fd\u0159i Nvidia se sna\u017eili to ud\u011blat tak, aby fantazie se stala realitou. <\/p>\n

Nov\u00fd v\u00fdvoj<\/a><\/noindex>, kter\u00fd se naz\u00fdv\u00e1 GauGAN, vytv\u00e1\u0159\u00ed n\u00e1dhern\u00e9 krajiny z velmi jednoduch\u00fdch skic (opravdu jednoduch\u00fdch - kruhy, \u010d\u00e1ry a to je v\u0161e). Samoz\u0159ejm\u011b, v z\u00e1kladu t\u00e9to technologie le\u017e\u00ed modern\u00ed technologie - konkr\u00e9tn\u011b generativn\u00ed adversari\u00e1ln\u00ed neuronov\u00e9 s\u00edt\u011b.
\n<\/a><\/noindex>
\nGauGAN umo\u017e\u0148uje vytv\u00e1\u0159et barevn\u00e9 virtu\u00e1ln\u00ed sv\u011bty - a nejen pro z\u00e1bavu, ale i pro pr\u00e1ci. Architekti, specialist\u00e9 na krajin\u00e1\u0159sk\u00fd design, v\u00fdvoj\u00e1\u0159i her - v\u0161ichni mohou \u010derpat n\u011bco u\u017eite\u010dn\u00e9ho. Um\u011bl\u00e1 inteligence okam\u017eit\u011b \u201ech\u00e1p\u011b\u201c co \u010dlov\u011bk chce a dopl\u0148uje p\u016fvodn\u00ed my\u0161lenku obrovsk\u00fdm mno\u017estv\u00edm detail\u016f. <\/p>\n

\u201eMozkov\u00fd storm v oblasti n\u00e1vrhu designu se daleko snadn\u011bji realizuje pomoc\u00ed GauGAN, proto\u017ee chytr\u00fd \u0161t\u011btec dok\u00e1\u017ee doplnit p\u016fvodn\u00ed skicu kvalitn\u00edmi obr\u00e1zky,\u201c uvedl jeden z v\u00fdvoj\u00e1\u0159\u016f GauGAN.<\/p>\n

U\u017eivatel\u00e9 tohoto n\u00e1stroje mohou m\u011bnit p\u016fvodn\u00ed zam\u00fd\u0161len\u00ed, modifikovat krajinu nebo jin\u00fd obr\u00e1zek, p\u0159id\u00e1vat nebe, p\u00edsek, mo\u0159e atd. V\u0161echno, co du\u0161i p\u0159eje, p\u0159i\u010dem\u017e p\u0159id\u00e1n\u00ed se realizuje za p\u00e1r sekund.<\/p>\n

Neuronov\u00e1 s\u00ed\u0165 byla tr\u00e9nov\u00e1na na z\u00e1klad\u011b milion\u016f obr\u00e1zk\u016f. D\u00edky tomu syst\u00e9m m\u016f\u017ee pochopit, co \u010dlov\u011bk chce a jak dos\u00e1hnout po\u017eadovan\u00e9ho. Nav\u00edc neuronov\u00e1 s\u00ed\u0165 nezapom\u00edn\u00e1 na drobn\u00e9 detaily. Tak\u017ee pokud nakresl\u00edte schematicky rybn\u00edk a n\u011bjak\u00e9 stromy vedle n\u011bj, po o\u017eiven\u00ed krajiny se v\u0161echny bl\u00edzk\u00e9 objekty odr\u00e1\u017eej\u00ed ve zrcadle vody rybn\u00edka.<\/p>\n

Syst\u00e9mu lze \u0159\u00edci, jak\u00e1 m\u00e1 b\u00fdt viditeln\u00e1 plocha - m\u016f\u017ee b\u00fdt pokryta tr\u00e1vou, sn\u011bhem, vodou nebo p\u00edskem. V\u0161echno to lze za sekundu p\u0159em\u011bnit, tak\u017ee sn\u00edh se stane p\u00edskem a m\u00edsto zasn\u011b\u017een\u00e9 pustiny dostane mal\u00ed\u0159 pou\u0161tn\u00ed krajinu. <\/p>\n

\u201eJe to jako omalov\u00e1nka, kter\u00e1 \u0159\u00edk\u00e1, kde um\u00edstit stromy, kde slunce a kde obloha. Pot\u00e9, po po\u010d\u00e1te\u010dn\u00edm \u00fakolu, neuronov\u00e1 s\u00ed\u0165 o\u017eiv\u00ed obraz, p\u0159id\u00e1 pot\u0159ebn\u00e9 detaily a textury, vykresl\u00ed odrazy. To v\u0161e vych\u00e1z\u00ed z re\u00e1ln\u00fdch obraz\u016f,\u201c \u0159\u00edk\u00e1 jeden z v\u00fdvoj\u00e1\u0159\u016f. <\/p>\n

\r\n \"P\u0159ehr\u00e1t\r\n
<\/div>\r\n <\/div><\/center>
\nI p\u0159esto, \u017ee syst\u00e9mu chyb\u00ed \u201epochopen\u00ed\u201c re\u00e1ln\u00e9ho sv\u011bta, vytv\u00e1\u0159\u00ed syst\u00e9m ohromuj\u00edc\u00ed krajiny. Je to proto, \u017ee zde pracuj\u00ed dv\u011b neuronov\u00e9 s\u00edt\u011b, gener\u00e1tor a diskrimin\u00e1tor. Gener\u00e1tor vytv\u00e1\u0159\u00ed obraz a ukazuje ho diskrimin\u00e1torovi. Ten na z\u00e1klad\u011b milion\u016f p\u0159edchoz\u00edch obraz\u016f vyb\u00edr\u00e1 nejrealisti\u010dt\u011bj\u0161\u00ed varianty. <\/p>\n

Pr\u00e1v\u011b proto gener\u00e1tor \u201ev\u00ed\u201c, kde by m\u011bly b\u00fdt odrazy. Je t\u0159eba poznamenat, \u017ee n\u00e1stroj je velmi flexibiln\u00ed a nab\u00edz\u00ed spoustu nastaven\u00ed. S jeho pomoc\u00ed lze tvo\u0159it obrazy, p\u0159izp\u016fsobovat se stylu ur\u010dit\u00e9ho um\u011blce, nebo se jenom hr\u00e1t s rychl\u00fdm p\u0159id\u00e1v\u00e1n\u00edm v\u00fdchodu nebo z\u00e1padu slunce. <\/p>\n

V\u00fdvoj\u00e1\u0159i tvrd\u00ed, \u017ee syst\u00e9m jednodu\u0161e nebere obr\u00e1zky odkudkoliv, neskl\u00e1d\u00e1 je dohromady a nedost\u00e1v\u00e1 v\u00fdsledek. Ne, v\u0161echny vznikl\u00e9 \u201eobr\u00e1zky\u201c jsou generovan\u00e9. Tedy neuronov\u00e1 s\u00ed\u0165 \u201etvo\u0159\u00ed\u201c, jako skute\u010dn\u00fd um\u011blec (nebo dokonce l\u00e9pe).<\/p>\n

Zat\u00edm nen\u00ed program voln\u011b dostupn\u00fd, ale brzy jej budete moci vyzkou\u0161et v akci. Toho se bude mo\u017en\u00e9 z\u00fa\u010dastnit na technologick\u00e9 v\u00fdstav\u011b GPU Technology Conference 2019, kter\u00e1 pr\u00e1v\u011b prob\u00edh\u00e1 v Kalifornii. \u0160\u0165astlivci, kte\u0159\u00ed mohli nav\u0161t\u00edvit v\u00fdstavu, ji\u017e mohou testovat GauGAN.<\/p>\n

Neuronov\u00e9 s\u00edt\u011b se u\u017e d\u00e1vno u\u010d\u00ed \u00fa\u010dastnit se tv\u016fr\u010d\u00edho procesu. Nap\u0159\u00edklad v lo\u0148sk\u00e9m roce n\u011bkter\u00e9 z nich mohly vytv\u00e1\u0159et 3D modely<\/a><\/noindex>. Krom\u011b toho v\u00fdvoj\u00e1\u0159i z DeepMind nau\u010dili neuronovou s\u00ed\u0165 obnovovat trojrozm\u011brn\u00e9 prostory a objekty podle kreslen\u00fdch obr\u00e1zk\u016f, fotografi\u00ed a n\u00e1\u010drtk\u016f. K vytvo\u0159en\u00ed jednoduch\u00e9 figury neuronov\u00e9 s\u00edti sta\u010d\u00ed jeden obr\u00e1zek, pro slo\u017eit\u011bj\u0161\u00ed objekty je pot\u0159eba p\u011bt obr\u00e1zk\u016f pro \u201enau\u010den\u00ed\u201c.<\/p>\n

Co se t\u00fd\u010de GauGAN, tento n\u00e1stroj jasn\u011b najde sv\u00e9 zaslou\u017een\u00e9 komer\u010dn\u00ed vyu\u017eit\u00ed \u2014 mnoho obor\u016f podnik\u00e1n\u00ed a v\u011bdy pot\u0159ebuje podobn\u00e9 slu\u017eby.<\/p>\n

Zdroj: habr.com<\/a><\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"

\u0412\u043e\u0434\u043e\u043f\u0430\u0434 \u043a\u0443\u0440\u0438\u043b\u044c\u0449\u0438\u043a\u0430 \u0438 \u0432\u043e\u0434\u043e\u043f\u0430\u0434 \u0437\u0434\u043e\u0440\u043e\u0432\u043e\u0433\u043e \u0447\u0435\u043b\u043e\u0432\u0435\u043a\u0430 \u0412\u0441\u0435 \u043c\u044b \u0437\u043d\u0430\u0435\u043c, \u043a\u0430\u043a \u043d\u0430\u0440\u0438\u0441\u043e\u0432\u0430\u0442\u044c \u0441\u043e\u0432\u0443. \u041d\u0443\u0436\u043d\u043e \u0441\u043d\u0430\u0447\u0430\u043b\u0430 \u043d\u0430\u0440\u0438\u0441\u043e\u0432\u0430\u0442\u044c \u043e\u0432\u0430\u043b, \u043f\u043e\u0442\u043e\u043c \u0435\u0449\u0435 \u043e\u043a\u0440\u0443\u0436\u043d\u043e\u0441\u0442\u044c, \u043d\u0443 \u0430 \u043f\u043e\u0442\u043e\u043c \u2014 \u043f\u043e\u043b\u0443\u0447\u0430\u0435\u0442\u0441\u044f \u0448\u0438\u043a\u0430\u0440\u043d\u0430\u044f \u0441\u043e\u0432\u0430. \u041a\u043e\u043d\u0435\u0447\u043d\u043e, \u044d\u0442\u043e \u0448\u0443\u0442\u043a\u0430, \u043f\u0440\u0438\u0447\u0435\u043c \u043e\u0447\u0435\u043d\u044c \u0441\u0442\u0430\u0440\u0430\u044f, \u043d\u043e \u0438\u043d\u0436\u0435\u043d\u0435\u0440\u044b Nvidia \u043f\u043e\u0441\u0442\u0430\u0440\u0430\u043b\u0438\u0441\u044c \u0441\u0434\u0435\u043b\u0430\u0442\u044c \u0442\u0430\u043a, \u0447\u0442\u043e\u0431\u044b \u0444\u0430\u043d\u0442\u0430\u0437\u0438\u044f \u0441\u0442\u0430\u043b\u0430 \u0440\u0435\u0430\u043b\u044c\u043d\u043e\u0441\u0442\u044c\u044e. \u041d\u043e\u0432\u0430\u044f \u0440\u0430\u0437\u0440\u0430\u0431\u043e\u0442\u043a\u0430, \u043a\u043e\u0442\u043e\u0440\u0430\u044f \u043d\u0430\u0437\u044b\u0432\u0430\u0435\u0442\u0441\u044f GauGAN, \u0441\u043e\u0437\u0434\u0430\u0435\u0442 \u0448\u0438\u043a\u0430\u0440\u043d\u044b\u0435 \u043f\u0435\u0439\u0437\u0430\u0436\u0438 \u0438\u0437 \u043e\u0447\u0435\u043d\u044c \u043f\u0440\u043e\u0441\u0442\u044b\u0445 \u043d\u0430\u0431\u0440\u043e\u0441\u043a\u043e\u0432 (\u0434\u0435\u0439\u0441\u0442\u0432\u0438\u0442\u0435\u043b\u044c\u043d\u043e […]<\/p>\n","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[],"tags":[],"class_list":["post-29895","post","type-post","status-publish","format-standard","hentry"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\n","aioseo_head_json":{"title":"\ud83e\udd47Neural Network od Nvidie prom\u011b\u0148uje jednoduch\u00e9 n\u00e1\u010drtky v kr\u00e1sn\u00e9 krajiny | ProHoster","description":"Vodop\u00e1d ku\u0159\u00e1ka a.","canonical_url":"https:\/\/prohoster.info\/cs\/blog\/nejroset-ot-nvidia-prevrashhaet-prostejshie-nabroski-v-krasivye-pejzazhi","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"cs_CZ","og:site_name":"ProHoster | \u041a\u0443\u043f\u0438\u0442\u044c \u043d\u0430\u0434\u0435\u0436\u043d\u044b\u0439 \u0445\u043e\u0441\u0442\u0438\u043d\u0433 \u0434\u043b\u044f \u0441\u0430\u0439\u0442\u043e\u0432 \u0441 \u0437\u0430\u0449\u0438\u0442\u043e\u0439 \u043e\u0442 DDoS, VPS VDS \u0441\u0435\u0440\u0432\u0435\u0440\u044b","og:type":"article","og:title":"\ud83e\udd47\u041d\u0435\u0439\u0440\u043e\u0441\u0435\u0442\u044c \u043e\u0442 Nvidia \u043f\u0440\u0435\u0432\u0440\u0430\u0449\u0430\u0435\u0442 \u043f\u0440\u043e\u0441\u0442\u0435\u0439\u0448\u0438\u0435 \u043d\u0430\u0431\u0440\u043e\u0441\u043a\u0438 \u0432 \u043a\u0440\u0430\u0441\u0438\u0432\u044b\u0435 \u043f\u0435\u0439\u0437\u0430\u0436\u0438 | ProHoster","og:description":"\u0412\u043e\u0434\u043e\u043f\u0430\u0434 \u043a\u0443\u0440\u0438\u043b\u044c\u0449\u0438\u043a\u0430 \u0438.","og:url":"https:\/\/prohoster.info\/cs\/blog\/nejroset-ot-nvidia-prevrashhaet-prostejshie-nabroski-v-krasivye-pejzazhi","og:image":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:secure_url":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:width":350,"og:image:height":350,"article:published_time":"2019-10-31T18:32:30+00:00","article:modified_time":"2019-10-31T18:32:30+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster","article:author":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":{"post_id":"29895","title":null,"description":null,"keywords":null,"keyphrases":null,"primary_term":null,"canonical_url":null,"og_title":null,"og_description":null,"og_object_type":"default","og_image_type":"default","og_image_url":null,"og_image_width":null,"og_image_height":null,"og_image_custom_url":null,"og_image_custom_fields":null,"og_video":null,"og_custom_url":null,"og_article_section":null,"og_article_tags":null,"twitter_use_og":false,"twitter_card":"default","twitter_image_type":"default","twitter_image_url":null,"twitter_image_custom_url":null,"twitter_image_custom_fields":null,"twitter_title":null,"twitter_description":null,"schema":{"blockGraphs":[],"customGraphs":[],"default":{"data":{"Article":[],"Course":[],"Dataset":[],"FAQPage":[],"Movie":[],"Person":[],"Product":[],"ProductReview":[],"Car":[],"Recipe":[],"Service":[],"SoftwareApplication":[],"WebPage":[]},"graphName":"Article","isEnabled":true},"graphs":[]},"schema_type":null,"schema_type_options":null,"pillar_content":false,"robots_default":true,"robots_noindex":false,"robots_noarchive":false,"robots_nosnippet":false,"robots_nofollow":false,"robots_noimageindex":false,"robots_noodp":false,"robots_notranslate":false,"robots_max_snippet":null,"robots_max_videopreview":null,"robots_max_imagepreview":"large","priority":null,"frequency":null,"local_seo":null,"seo_analyzer_scan_date":"2026-01-20 22:55:20","breadcrumb_settings":null,"limit_modified_date":false,"reviewed_by":null,"ai":null,"created":"2021-03-01 03:46:27","updated":"2026-01-20 22:55:20","focus_keyword":null,"additional_keywords":null,"truseo_locale":null},"gt_translate_keys":[{"key":"link","format":"url"}],"_links":{"self":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/posts\/29895","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/comments?post=29895"}],"version-history":[{"count":0,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/posts\/29895\/revisions"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/media?parent=29895"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/categories?post=29895"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/tags?post=29895"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}